1
\$\begingroup\$

I am using TPS63001 Buck-Boost regulator for creating 3.3V in my circuit. The converter works in Buck mode as excepted. So when I give 5V on the input, it is OK. When I reduce the input voltage from 5V to 2.7V, it still works perfectly in this case in Boost mode. But when I turn the input voltage off, and try to start the circuit from 2.7V, it does not work. It consums ~200 mA current, and gives a buzzing noise out, the output voltage is not 3.3V

My circuit schematic is following: enter image description here The EN is connected in this case to VCC. I tried to change the inductor, but the same effect is shown. What could be the problem?

EDIT

I measured the EN input voltage, and it falls to ~1.6V, so the chip can't turn on. The VCC is created following:

enter image description here

The Forward Voltage of the Schottky diodes are maximal ~0.4V. So if the system gets 2.7V, the VCC should be at least 2.3V, but it falls down to 1.6V. If I use a wire to connect 2.7V to EN directly, the konverter turns on and remains turned on when I remove the wire.

The layout is OK, I followed the recommendations in the datasheet. There is currently no load on the output, because I want to solder up the microncotroller, when the power supply is stable. The chip should be fine, because I have 2 boards with the same chip and same layout and the same effect is shown. Can the Schottky diode be problem? There is 1V droping on the diode, if the regulator is working fine, than it is ~0.35V.

EDIT

The layout is following:

enter image description here

The C1 is close to the Vin.

\$\endgroup\$
11
  • 1
    \$\begingroup\$ In the datasheet you posted pin 6 is enable and pin 7 is ps also your part above has two pin sixes are you sure you have everything hooked up correctly? \$\endgroup\$
    – Some Hardware Guy
    Commented Oct 12, 2015 at 13:09
  • 1
    \$\begingroup\$ Looks like he's treated the power pad as pin 6, adding 1 to every pin greater than 5. \$\endgroup\$
    – michaelyoyo
    Commented Oct 12, 2015 at 13:19
  • \$\begingroup\$ Yes, the pinout is correct, I reviewed it. The pin 6 is the power pad. \$\endgroup\$
    – Milan Tenk
    Commented Oct 12, 2015 at 13:20
  • \$\begingroup\$ What kind of battery is this? You need to figure out why the voltage drops, measure it at all available points and measure the resistance of the battery and of the path until VCC. Also where exactly is C1 located in respect to D1/D2/S1/BAT etc. \$\endgroup\$
    – PlasmaHH
    Commented Oct 12, 2015 at 14:12
  • \$\begingroup\$ Currently I'm not using battery, I use an adjustable power supply with power cut. I measure them and post the results soon. \$\endgroup\$
    – Milan Tenk
    Commented Oct 12, 2015 at 14:15

2 Answers 2

Reset to default
2
\$\begingroup\$

The data sheet states that the minimum start-up voltage is 1.9 volts so it can only be a couple of things that are causing this problem: -

  • Your input voltage is dropping below 1.9 volts under the heavy load conditions on a cold restart
  • The line/wire impedance to your converter from the power supply has too much resistance/inductance
  • Maybe it's a PCB layout problem (any breadboard layout will be problematic for this type of design)
  • Your measurement of the input voltage is incorrect.
  • Bad chip.

So go thru this list and try and figure out what is the problem. Double check the circuit you have shown - I see discrepancies with the data sheet - pin 10 is FB not pin 11. Pin 9 is ground in the data sheet yet you have pin 9 as VIN A. Most strange.

\$\endgroup\$
5
  • \$\begingroup\$ There is a problem with the EN voltage. I have edited my question, I went trough your list. The pinout is OK. \$\endgroup\$
    – Milan Tenk
    Commented Oct 12, 2015 at 14:08
  • \$\begingroup\$ Where is C1 - is it closely attached to the chip's Vin pin? Are you using a circuit board or some prototyping method? \$\endgroup\$
    – Andy aka
    Commented Oct 12, 2015 at 14:33
  • \$\begingroup\$ C1 is closely attached, I added a layout schreenshot to the question. I'm using a PCB board with 2 layers. \$\endgroup\$
    – Milan Tenk
    Commented Oct 12, 2015 at 14:45
  • \$\begingroup\$ I'm not sure you have a decent enough PCB layout for this chip. Have you read the data sheet section on page 16 - capacitors do not appear to be as close to the chip as possible. \$\endgroup\$
    – Andy aka
    Commented Oct 12, 2015 at 15:04
  • \$\begingroup\$ You are right, it could be more compact. Next time i will consider this more seriosly. I posted the solution as answer. \$\endgroup\$
    – Milan Tenk
    Commented Oct 12, 2015 at 15:41
1
\$\begingroup\$

I found the solution for this problem. The forward current of the original Schottky diode (D2) was 120 mA, because the estimated current consumption of my system was ~70mA. It seems, that the power supply needs more for a short time, and thats why it could not turn on. After replacing the Schottky diode, the power supply works well in every case.

Thank all of you for the help!

\$\endgroup\$

Not the answer you're looking for? Browse other questions tagged or ask your own question.